Hyperreflective Dots on OCT as a Predictor of Treatment Outcome in Diabetic Macular Edema: A Systematic Review

Haifan Huang, Nomdo M. Jansonius, Haoyu Chen, Leonoor I. Los*

*Corresponding author for this work

Research output: Contribution to journalReview articlepeer-review

5 Citations (Scopus)
55 Downloads (Pure)


Topic: This review aims to evaluate the role of hyperreflective dots (HRDs), detected using OCT, as a predictor of the treatment outcome in patients with diabetic macular edema (DME).

Clinical relevance: The treatment of DME is possible, but its results are often unsatisfactory. Thus, it is important to develop biomarkers that can help to predict the treatment response to optimize the treatment's effect for individual patients.

Methods: PubMed, Embase, Web of science, and Cochrane library were searched (final search date on May 5, 2021). Participants were patients diagnosed with DME and provided with treatment. The predictor was HRDs, detected using OCT, before treatment. The outcomes were best-corrected visual acuity (BCVA) and central macular thickness (CMT), detected using OCT, after treatment. Two reviewers independently screened the titles and abstracts as well as full text. The refined Quality in Prognosis Studies tool was used to assess the risk of bias for each included study. Because of the clinical heterogeneity of the studies, a meta-analysis was not performed.

Results: Thirty-six studies were included. The Quality in Prognosis Studies assessment showed that most studies had a low or moderate risk of bias in 6 domains. Six studies could not find any correlation between baseline HRDs (either the presence or absence of HRDs [n = 1] or baseline HRD number [n = 5]) and outcome (BCVA or CMT), whereas 12 studies found a significant correlation between these variables. Eight studies reported that baseline HRDs could predict a poor visual outcome (n = 4 on prescence or abscence of HRD and n = 4 on HRD number), and 4 studies (n = 1 on prescence or abscence of HRD and n = 3 on HRD number) found that HRDs were predictive of visual improvement. Fifteen out of 17 studies found that the HRD number decreased after treatment.

Conclusion: Based on the current literature, the HRD numbers decrease with treatment, but it is not clear whether HRDs predict the treatment outcome in patients with DME. Future investigations with more uniform approaches are needed to confirm the nature of this biomarker and its effect on DME treatment outcome.

Original languageEnglish
Pages (from-to)814-827
Number of pages14
JournalOphthalmology Retina
Issue number9
Publication statusPublished - Sept-2022


  • Diabetic macular edema
  • Diabetic retinopathy
  • Hyperreflective dots
  • OCT
  • Systematic review

Cite this